Breast Cancer diagnosis using, grey-level co-occurrence matrices, decision tree classification and evolutionary feature selection

Breast Cancer is the most widespread Cancer among women. Breast cancer is the second leading cause of cancer death in women. The number of new cases of breast cancer was 124.8 per 100,000 women per year. The number of deaths was 21.9 per 100,000 women per year. These rates are age-adjusted and based on 2008-2012 cases and deaths. This represents about 12% of all new cancer cases and 25% of all cancers in women. Conventional diagnosis methods of Breast Cancer include biopsy, mammography thermography, and Ultrasound imaging. Among these methods, mammography is the most efficient method for the early diagnosis of Breast Cancer. Detecting Breast Cancer and classifying mammography images are the standard clinical procedures for the diagnosis of Breast Cancer. In order to classify mammography, is provided automated computer-based detection methods. In this study, Gray-Level Co-occurrence Matrix and Cumulative Histogram features were used. We also use a Decision Tree as a classifier system. Then we introduce a new algorithm that called "Discrete Version of Imperialist Competitive Algorithm" as a global optimization algorithm in discrete space, and we use this algorithm for finding the best features of the extracted features.